INDEX

Explanations

documentation and guides

This neuron fires on tokens that are part of technical documentation URLs, especially domain/path segments (like “php.net/manual”) and their numeric version or section identifiers.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

the

-1.59

 that

-1.45

 because

-1.39

 čís

-1.31

We

-1.27

 fără

-1.20

__':

-1.19

ANIEL

-1.18

kDa

-1.17

 eftersom

-1.16

POSITIVE LOGITS

 renova

1.40

 malang

1.33

bbene

1.31

 custodia

1.27

 fatis

1.26

 veterin

1.25

 ruine

1.24

 bogor

1.23

 formen

1.22

řel

1.22

Activations Density 0.285%